The work described in this application is aimed at preparing optical sensors with imaging fibers for creating platforms that can be used for high throughput and high content saliva analysis. This application describes experiments aimed at ascertaining whether both specific and non-specific multi-analyte sensor arrays can be used for distinguishing various disease phenotypes. With some, and perhaps most, diseases, it is unlikely that there is a specific compound present that signals the presence of disease. Instead, it is likely that the disease is manifested by a pattern of compounds including both common metabolites possibly at altered levels, or unusual metabolites, all in the context of a complex and non-constant background. It is the intention of the present work to explore the possibility that sensor arrays can be designed to recognize these metabolic patterns in changing backgrounds (e.g. different patients, different secondary diseases, different drug regimens). The major goals of the application are to develop testing protocols, and to prepare sensors with sufficient diversity and sensitivity to make such measurements. This application focuses on the first developments to use saliva as a diagnostic tool for the diagnosis of oral and systemic disease states using micro-array sensing systems. Consequently, it is imperative to evaluate the response pattern obtained under a variety of conditions. Such conditions must take into account both the unique physiological parameters of salivary secretions and whole saliva as well as the application of different micro-sensor technologies and methods. The arrays will be integrated into a sample preparation and readout system with the ability to process 96-1536 samples at a time. The small size of the samples required for detection minimizes sample preparation time and reduces reagent storage and handling requirements. The integrated system should have the highest throughput in terms of both high sample numbers and processing/readout time of any system. This application describes initial work in a long-term effort to develop a new approach to using saliva for diagnostic purposes. The overall goal is to develop both new tools and a universal platform for performing measurements on saliva samples. Much of the initial work will be directed at screening potentially useful sensing chemistries for their applicability to sensing components in saliva.